root d37bfa9d75 [update] harden training and inference reliability ... Fix truncated SFT/DPO label masking and resume-safe gradient accumulation to prevent unstable optimization, and add explicit eval backend selection plus regression checks for safer local validation.		2026-04-05 18:11:37 +00:00
..
rollout_engine.py	[update] minimind-3	2026-03-25 23:57:45 +08:00
train_agent.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_distillation.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_dpo.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_full_sft.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_grpo.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_lora.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_ppo.py	[update] minimind-3	2026-03-25 23:57:45 +08:00
train_pretrain.py	[update] harden training and inference reliability	2026-04-05 18:11:37 +00:00
train_tokenizer.py	[update] minimind-3	2026-03-25 23:57:45 +08:00
trainer_utils.py	[update] minimind-3	2026-03-25 23:57:45 +08:00